Path-Based Statistical Modeling of Multipath Components in Propagation Channels for Wireless Communications in Unmanned Aviation

Unmanned aviation (UA), including both small drones in urban airspace, as well as larger unmanned airplanes, is one of the most popular topics in aviation these days. One of the key enablers for unmanned aviation is a secure and robust communication link between the air vehicle and the instance controlling and/or monitoring the air vehicle, e.g. a remote pilot. Naturally, parts of this communication link are wireless; here, we assume a terrestrial link between the air vehicle’s radio and the ground stations and a vehicle-to-vehicle communication link between drones.All wireless communication is subject to certain channel effects, e. g. multipath propagation, that usually degrade the radio signal during transmission. A good understanding of these channel effects is of high importance during the development of new wireless waveforms. A common approach to gain knowledge on the characteristics of a wireless channel is to perform channel measurements: The DLR performed several channel measurement campaigns involving smaller drones and a jet aircraft in the recent years. The collected data contain information on the channel characteristics during the respective scenarios. In this paper, we focus on the detection, path-based tracking, and modeling of multipath components and their evolution over time. First, we present our processing chain for the detection and tracking of multipath components and apply it to the collected measurement data. We then introduce a compact representation of the evolution of the detected multipath components over time. The statistical properties of this representation are then used to fit a kernel that is used to generate artificial multipath-components and their evolution. We finally evaluate our approach by comparing the delay spread and the K-factor of the measurement data with the corresponding properties of the data generated by our model.